Hemoglobins are widely but sporadically distributed in most invertebrate phyla. Most of these hemoglobins have 15-17 Kd chains and are either intracellular (molecules less than or equal to 65Kd) or extracellular with molecular weights to several million. Some arthropods and molluscs have hemoglobins which have chains composed of 2 to 20 myoglobin-like units joined by peptide bonds. Many of the hemoglobins exhibit highly cooperative O2 binding and are strongly affected by pH. Others are non-cooperative and completely independent of pH. This extraordinary diversity of both form and function provides an attractive system in which to investigate gene structure and evolution. We plan to determine the cDNA and gene structure of globins from 4 invertebrates: Urechis (intracellular, about 14Kd chains), Lumbricus (extracellular, 15-18Kd chains), Barbatia (intracellular, about 33Kd chains with 2 heme-containing domains and 17Kd chains), and Cardita (extracellular, 290Kd Chains each with about 20 heme-domains). We will learn if, as seems most unlikely, the domains arose by unique RNA-processing events. If the genes of the domain globins show the expected repeating structure, then the differences between them should indicate a pattern of duplication. Existing data suggest that the about 20 domains in Cardita globin may all be very similar. The gene structure should allow assessment of the possible roles of unequal crossing-over and gene conversion. The leghemoglobin gene of leguminous plants has 3 introns and 4 exons in contrast to the vertebrate globin gene which has 2 introns. At this time the only known structure of an invertebrate globin gene is that of the larval insect, Chironomus, but this gene has no introns so sheds no light on the nature of the ancestral globin gene with regard to the numbers of introns. We have determined the amino acid sequence of one of the 8 kinds of chain of the binuclear copper protein, hemocyanin, from the horseshoe crab, Limulus. We propose to determine the structures of the remaining 7 chains by recombinant DNA techniques and to use the cDNA as a probe for the hemocyanin genes. Hemocyanins have been shown to be homologous with Neurospora tyrosinase. We plan therefore to isolate tyrosinase, if possible, from Limulus and/or other arthropods and molluscs known to possess both tyrosinase and hemocyanin. We will isolate the poly(A)+RNA, and prepare cDNA which will be cloned and screened with appropriate oligonucleotides. We will then use the cDNA to screen genomic libraries for the tyrosinase and hemocyanin genes.